The ability to select proper actions is critical for flexible adaptive behavior. In vertebrates, multiple neural systems have evolved to coordinate different aspects of motor selection, execution, and learning. Key among these systems is the basal ganglia, a set of subcortical nuclei that are critical for motor planning and habit learning, and which are also implicated in Parkinson disease (PD), among the most commonly-diagnosed movement disorders. The anatomical connectivity of the basal ganglia is well characterized, and general hypotheses about basal ganglia function and the role of dopamine have been proposed. However, the mechanisms underlying dysfunction of basal ganglia circuits in PD is not well understood. In order to gain new insight into the dysfunction of the basal ganglia, we will target two new understudied aspects of basal ganglia circuitry: (1) the inputs from the thalamus to the striatum, and (2) the outputs from the substantia nigra reticulata/entopeduncular nucleus to the pedunculopontine tegmentum (PPTg). We have obtained compelling preliminary evidence that these regions are involved in motor dysfunction in animal models of PD, and we propose to thoroughly investigate the cellular, synaptic, and circuit mechanisms that underlie this dysfunction. The goal is to identify new therapeutic targets and strategies for treating PD, without the debilitating side effects associated with long-term use of dopamine replacement therapy.